Post on 27-Nov-2020
Project ReportATC-244
Anomalous Propagation Ground Clutter
Suppression with the Airport SurveillanceRadar (ASR) Weather Systems Processor (WSP)
J. A. Cullen
15 March 1996
Lincoln Laboratory MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LEXINGTON, MASSACHUSETTS
Prepared for the Federal Aviation Administration, Washington, D.C. 20591
This document is available to the public through
the National Technical Information Service, Springfield, VA 22161
This document is disseminated under the sponsorship of the Department of Transportation in the interest of information exchange. The United States Government assumes no liability for its contents or use thereof.
1. Report No.
ATC-244
2. Government Accession No.
TECHNICAL REPORT STANDARD TITLE PAGE
3. Recipient's Catalog No.
,
4. Title and Subtitle
Anomalous Propagation Ground Clutter Suppression with the Airport SurveillanceRadar (ASR) Weathel' Systems Processor (WSP)
7. Author(s)
Joseph A. Cullen
9. Performing Organization Name and Address
Lim:oln Lahoratory, MIT244 Wood Stl'eetLexington, MA02173-9108
12. Sponsoring Agency Name and AddressDepartment of TI'anspOl·tationFederal Aviation AdministrationWashington, DC 20591
15. Supplementary Notes
5. Report Date15 March 1996
6. Performing Organization Code
8. Performing Organization Report No.
ATC-244
10. Work Unit No. (TRAIS)
11. Contract or Grant No.
DTFAOl-93-Z-02012
13. Type of Report and Period Covered
Project Report
14. Sponsoring Agency Code
This l'ellOl't is hased on studies performed at Lincoln Lahoratory, a center for research operated hy Massachusetts Institute ofTedmoloh'Y umler Ail' FOl'ce Contract F19628-95-C-0002.
16. Abstract
Ground-dutter hreakthrough caused hy anomalous propagation (AP)-ducting of the radar heam whenpassing through significant atmospheric temperature and/or moisture gradients-is a significant issue for airtraffi,: controllers who use Airport Surveillance Radar (ASR) weather channel data to guide aircraft throughthe airport terminal area. At present, these data are often contaminated with AP, leaving the controllerunsure alHlut the validity of information on storm location and intensity.
The Weathet, System Processor (WSP), which is scheduled for deployment at 33 airports in the U.S.,indudes an AP-Editing algorithm designed to remove AP hased on its Doppler-spectrum characteristicsin ASR-9 data. This report provides a description of the algorithm currently used in the FAA/LincolnLahoratory WSP prototype and a measurement of the performance of the algorithm during nine episodes ofAP anll/or true weather in Orlando, Florida in 1991 and 1992.
17. Key Words
Anomalous PropagationGround Clutter BreakthroughSpectral DiscriminationASR Weathet, System PrOl:essor (WSP)
18. Distribution Statement
This document is availahle to the puhlic through theNational Technical Information Service,Springfield, VA 22161.
19. Security Classif. (of this report)
Undassifie.l
20. Security Classif. (of this page)
Undassified
21. No. of Pages
28
22. Price
FORM DOT F 1700.7 (8-72) Reproduction of completed page authorized
ABSTRACT
Ground-elutter breakthrough caused by anomalous propagation (AP}----ducting of the radarbeam when passing through significant atmospheric temperature and/or moisture gradients-is asignificant issue for air traffic controllers who use Airport Surveillance Radar (ASR) weatherchannel data to guide aircraft through the airport terminal area. At present, these data are oftencontaminated with AP, leaving the controller unsure about the validity of information on stormlocation and intensity.
The Weather System Processor (WSP), which is scheduled for deployment at 33 airports in theU.S., includes an AP-Editing algorithm designed to remove AP based on its Doppler-spectrumcharacteristics in ASR-9 data. This report provides a description of the algorithm currently used inthe FAA/Lincoln Laboratory WSP prototype and a measurement of the performance of thealgorithm during nine episodes of AP and/or true weather in Orlando, Florida in 1991 and 1992.
III
ACKNOWLEDGMENTS
The authors are especially grateful to Evelyn Mann for modifications to her AP-Edit scoring andoff-line processing software that made this evaluation possible. Thanks also to Oliver Newell forproviding information about the WSP AP-Editing Algorithm.
v
TABLE OF CONTENTS
Section
Abstract
Acknowledgments
List of Illustrations
List of Tables
1. INTRODUCTION
iii
v
IX
IX
1
2. FAA / LINCOLN LABORATORY WSP PROTOTYPE DESCRIPTIONAND AP-EDITING APPROACH 3
3. TEST CASES AND SCORING METHODOLOGY 5
4. PERFORMANCE STATISTICS 7
5. COMPARATIVE RESULTS 11
6. CONCLUSION AND RECOMMENDATION 13
GLOSSARY 15
REFERENCES 17
Vll
LIST OF ILLUSTRATIONS
Figure Page
1. ASR-9 data showing visual difference between anomalous propagation andtrue-weather echoes. 6
2. Unedited (left) and corresponding AP-edited data (right) from 7/10/92. 9
3. WSP high-level architecture. 13
LIST OF TABLES
1. Time Periods Used For WSP AP-Editing Algorithm Evaluation
2. AP-Editing Statistics: WSP AP-Editing Algorithm
3. Simulated ITWS AP-Editing Results
4. Comparison of ITWS and WSP AP-Editing Results
IX
5
7
11
11
1. INTRODUCTION
Current Airport Surveillance Radars (ASR-9s) [Taylor and Bronins, 1985] feature a dedicateddigital processor that detects and displays six calibrated levels of precipitation reflectivity on terminal Air Traffic Control (ATC) radar scopes. The following parameters of the ASR-9 and its "weatherchannel" make this system well suited to provide storm location and intensity information for terminal ATC operations:
1. Operation at a non-attenuating 10 cm wavelength,
2. A vertically integrating, cosecant-squared antenna elevation pattern that detects precipitation echoes over the entire altitude interval of concern for terminal operations,
3. An update rate (30 seconds) consistent with the tight temporal and spatial tolerances of terminal ATC, and
4. Real-time display of precipitation reflectivity on the same Data Entry and Display System (DEDS) and Bright Radar Indicator Tower Equipment (BRITE)scopes that radar and tower controllers employ for monitoring aircraft position.
In view ofthese attributes, the Federal Aviation Administration's (FAA) National Airspace System (NAS) development plan has always considered ASR-9 weather channel as the primary sourceof storm reflectivity information in terminal airspace, even at airports that are equipped with the Terminal Doppler Weather Radar (TDWR) [Evans and Turnbull, 1989] for low-altitude wind sheardetection. ASR-9 weather channel data will likewise be a key input to the Integrated TerminalWeather System (lTWS) [Klingle-Wilson, 1995] which will be deployed at major U.S. airports toprovide comprehensive, user-oriented information about weather conditions affecting terminal airspace.
Widespread operational use of the ASR-9 weather channel has revealed one significant issue:ground clutter breakthrough caused by anomalous propagation of radar signals (AP) - superrefractive propagation in an environment characterized by strong, low-altitude moisture and/or temperature gradients. After processing by the weather channel's aggressive spatial and temporal smoothing[Weber, 1986], this clutter breakthrough is largely indistinguishable from actual meteorologicalechoes. As described in Battan, 1973, such breakthrough can occur both during widespread temperature inversions and as a result of thunderstorm outflows ofmoist, rain-cooled air. The latter is particularly stressing operationally since the AP-induced "false weather" coexists with actual thunderstorms that are affecting terminal area operations. Numerous "unsatisfactory condition reports"(UCRs) have been filed by air traffic controllers in response to this problem.
The Weather System Processor (WSP) [Weber and Stone, 1995] is a developmental add-on tothe ASR-9 that provides detailed automated information on precipitation reflectivity, storm motion,and associated low-altitude wind shear by processing received echoes for meteorological Dopplercontent. Where implemented, the WSP will assume the functionality ofthe current ASR-9 six-levelweather channel, providing six-level weather reflectivity images to both existing controller scopes(DEDS and BRITE) and a color Geographic Situation Display (GSD) intended for use by supervisory personnel and traffic management specialists. Analysis of the Doppler spectrum and "high" versus "low" receiving beam relative amplitudes of received echoes provides a mechanism for the WSP
to discriminate between true meteorological echoes and AP-induced ground clutter breakthrough.This capability has been demonstrated operationally on the Lincoln Laboratory / FAA WSP prototype in Orlando, FL and Albuquerque, NM with good success.
The current FAA NAS architecture calls for deployment of the WSP only at smaller airportsnot equipped with a TDWR. At the larger TDWR-supported airports, AP-editing will be accomplished by ITWS, which compares ASR six-level reflectivity data with reflectivity data from otherweather radars to identify and edit regions of AP contamination [Klingle-Wilson, et aI., 1995]. Deficiencies in this technique can occur when the radar used for comparison also experiences AP or whenAP conditions are changing rapidly, as relevant data from other radars may be several minutes old.Furthermore, a "cone-of-silence" exists around the corroborating radars due to scanning strategiesthat do not allow full sampling of nearby storms. Therefore, because of the possibility of editingtrue-weather echoes, AP editing cannot be performed within this region. Finally, ITWS does notprovide AP-edited precipitation on controller's DEDS and BRITE displays. As a result, supervisorypersonnel at the 45 TDWR-equipped airports where ITWS is planned for deployment will have toverbally relay AP information from their GSD to controllers. This arrangement will increase controller workload.
This report describes the WSP AP-Editing Algorithm and evaluates its performance using datacollected during prototype WSP operations in Orlando, FL - an environment subject to frequent andintense episodes ofanomalous propagation. Section 2 describes the current FAA / Lincoln Laboratory WSP prototype and its AP-editing approach. Section 3 summarizes the data set used to score thealgorithm and describes the scoring exercise. Section 4 lists performance results and Section 5compares these with off-line ITWS (Integrated Terminal Weather System) AP-Editing results forsome of the same test cases. In Section 6, we conclude the report by describing a modular WSP architecture that would allow for relatively low-cost implementation across the NAS of the subset ofWSP hardware and software required to generate AP-free precipitation reflectivity information.
2
,.
2. FAA I LINCOLN LABORATORY WSP PROTOTYPE DESCRIPTIONAND AP-EDITING APPROACH
The Weather System Processor is an outboard radar receiving channel and data processing computer, sharing the transmitter, timing signal generation and antenna subsystems of the ASR-9. Ahigh speed vector-oriented processor performs signal input, interference suppression and base-datageneration. The base data generated by these operations-precipitation reflectivity (DZ), radial velocity (V), spectrum width (SW) and signal-t<r-noise ratio (SNR)-are then reformatted by an output processor and distributed to meteorological detection algorithms. These algorithms are implemented by single-board computers or UNIX workstations, and the resulting graphical and/oralphanumeric output is sent to remote workstations and monitors in the air traffic control tower foruse by controllers and their supervisors [Weber, 1992].
The signal processing algorithms suppress ground clutter using techniques that are analogousto those employed by the ASR-9's six-level weather processor [Weber, 1987]. "Clear Day Maps"(CDM) are constructed that encode-for each range-azimuth resolution cell-the mean residualclutter power output from each of three separate 17-point finite impulse response (FIR) high-passfilters as well as the mean unattenuated clutter level. The three filters achieve suppression of nominally 16,31 and 45 dB with associated stop-band widths (at the 3-dB down points) of2.8, 4.7 and6.1 mis, respectively. For each range gate, the least attenuating clutter filter (or an all-pass function)is chosen that permits output signal power to exceed the corresponding mean clutter residue levelin the CDM by a threshold (nominally 10 dB). This method minimizes the possibility of reflectivityestimate bias due to filtering of low-Doppler precipitation echoes. Unfortunately, as with the existing six-level weather processor algorithm, AP-induced ground clutter will often be passed throughthe WSP's clutter suppression module unattenuated since it is not accurately represented by theCDM.
Following clutter suppression, the signal processing algorithms generate DZ, V, SW and SNRproducts as follows. Autocorrelation function estimates for lags varying from zero to four times themean pulse repetition interval are computed for both beams and used to generate weather echo spectrum moments. Precipitation reflectivity factor and signal-t<r-noise power ratio are computed usingthe low-beam data and a pre-determined estimate of system noise level in each beam. The phaseangle of the low-beam estimate provides the Doppler-velocity estimate for the gust front algorithm.Low- and high-beam Doppler-velocity estimates are combined to create a surface velocity field forthe microburst detection algorithm [Weber, 1989]. Finally, spectrum width in the low beam is generated by applying a weighted, quadratic regression to the logarithms ofthe magnitudes of the autocorrelation function estimates.
Puzzo, et al. [1989] suggest various techniques for identifying and censoring AP in ASR-9 data.The method employed in the current WSP prototype involves 1) flagging all range bins with spectrum-width value less than 0.7 mls as AP, as well as neighboring range bins within a "window" offive range bins, centered on the selected bin, and 2) reducing the possibility of editing true weatherby removing the flag for range bins containing Doppler-velocity values greater than 1.0 mis, evenif they lie within the five-bin window mentioned above. These spectrum-width and Doppler-velocity thresholds were chosen based on expected spectral characteristics of clutter breakthrough (e.g.,antenna-scan modulation width) and testing with actual data. The current technique and individualparameters employed by the algorithm remain under scrutiny and may be modified. For example,
3
visual comparisons between simultaneous low and high-beam ASR-9 data indicate signal returnfrom the former exceeds that from the latter by 10 dB or more during AP episodes, with larger differences prominent during severe AP cases. This information could also be incorporated into the WSPAP-Editing Algorithm as an additional discriminant. Thus, the results reported here should beviewed as the lower limit of the AP-editing success that is achievable in the WSP.
4
3. TEST CASES AND SCORING METHODOLOGY
The database used to evaluate performance of the WSP AP-Editing algorithm was limited totime periods when both unedited and AP-edited files were archived. This restricted possible testcases to the late summer of 1991 and early summer of 1992 when the ASR-9 testbed was locatedin Orlando, FL. Furthermore, data were not routinely collected during late evenings or early mornings when nocturnal inversions often induce AP. The main source of AP episodes, therefore, wascool moist thunderstorm outflows (gust fronts) that passed over the radar site. These episodes typically lasted more than 30 minutes and sometimes a few hours. Table 1 shows the time periods thatwere identified after an extensive search through the candidate data and subsequently used to evaluate WSP AP-Editing algorithm performance. Three time periods featuring no AP (true weatherechoes only) and two periods containing no true weather echoes (AP only) were included in the testset. (The two time periods from 8/9/91 provided one example of each.)
Table 1.Time Periods Used For WSP AP-Editing Algorithm Evaluation
DAY OF DATA TIME PERIOD (UT)
8/2/918/9/918/10/917/10/927/13/927/15/927/17/927/18/927/20/92
2055-22252055-2145,2210-23002035-22001930-20451910-20100110-01301930-21002145-22101850-1950
ASR-testbed reflectivity data in 5-dBZ increments and at one or twcrminute intervals fromthese time periods were examined to generate AP truth. The presence of AP was clearly discernablewhen viewed in this manner because of its highly mottled appearance in contrast to true weatherechoes which exhibited smooth transition between dBZ levels (Figure 1). ASR low-beam velocityfields and Terminal Doppler Weather Radar (TDWR) testbed data served as additional guides in thetruthing process, as AP returns are usually associated with zercrDoppler velocity and the TDWRwas not as susceptible to AP as the ASR. Locations of AP with respect to the radar location werewritten to data files by using interactive display software that allowed the "truther" to enclose withinpolygons any AP regions on a particular scan. Then an automated scoring program which comparedunedited and AP-edited display files on a pixel-by-pixel basis ascertained whether AP and/or trueweather pixels were correctly/incorrectly edited by referring to the corresponding AP truth file. Thescoring software generated cumulative counts for the total number of AP and true-weather pixelswithin the range considered (28.8 kIn from ASR), and of the number correctly/incorrectly edited,to yield Probability of Editing AP (PEAP) and Probability of Editing Weather (PEW) statistics. Thescoring program also provided an option to ignore regions ofcontiguous pixels having total area lessthan a specified value. For this evaluation, pixels comprising contiguous-pixel regions of area lessthan 1.0 square kilometer were not considered since these small AP patches likely would not be perceived as significant weather by an operational user of the data.
5
Fixure I ASR-CJ data showinX visual difference between anomalous propagation and true-weather echoes. Ragged echoes to the west through northwest andnortheast through sOlllh of the radar are caused by AP. while smooth echoes southwest of the radar represent actual precipitation. These \'isual differences are notapparent to U.H'r.\ ofASR-9 data bemuse it is presented in VIP six-level format and smoothed extensively.
... ..
4. PERFORMANCE STATISTICS
Two of the nine data cases (8/10/91 and 7/17/92) had unedited and AP-edited scans that differedby approximately thirty seconds. Initial scoring of these cases revealed that scan-ta-scan variationin reflectivity values associated with true weather was too significant to obtain reliable PEW valuesbecause any change in video integrator processor (VIP) level for a particular pixel location wasscored as an "edit." For example, if displacement of a Level-2 weather pixel by a Level-l pixel wascaused by movement of weather during the 30-second interval or by statistical fluctuations in echostrength between the unedited and AP-edited scans, the pixel was deemed edited. This problem wasnot as pronounced in corresponding PEAP statistics, however, because AP echoes exhibit greatertemporal stability. Table 2 summarizes PEAP and PEW statistics obtained from scoring the test dataset. For each case, PEAP is expressed as the ratio of AP pixels that were (correctly) edited to thetotal number of AP pixels, while PEW is expressed as the ratio of true-weather pixels that were (incorrectly) edited to the total number of true-weather pixels.
Table 2.AP-Editing Statistics: WSP AP-Editing Algorithm
PIXELS > VIP Level 3 PIXELS > VIP Level 2
DATE PEAP % PEW % PEAP % PEW %
8/2/91 1755/1889 93 24/3431 0.7 16,445/17,414 94 579/15,883 3.6
8/9/91 * 17,014/17,640 96 0/2622 0 59,688/64,799 92 22/11,336 0.2
8/10/91 4047/4290 94 NA - 22,550 / 24,525 92 NA -
7/10/92 215/225 96 10/434 2.3 6012/6784 89 306/3473 8.8
7/13/92 noAP - 94/15,859 0.6 noAP - 437/46,066 0.9
7/15/92 noAP - 140/30,918 0.5 noAP - 709/72,944 1.0
7/17/92 5914/6024 98 NA - 30,091 /30,721 98 NA -7/18/92 913/950 96 0/358 0 5573/5896 95 51 /2240 2.3
7/20/92 1160/1261 92 - - 7168/7713 93 0/108 0
31,018/32,279 268/53,622 147,527/157,852 2104/ 152,050TOTALS (96.1%) (0.5%) (93.5%) (1.4%)
* PEAP statistics are from 2055-2145 time period; PEW statistics are from 2210-2300
PEAP for the individual cases varied from 92-98 percent for VIP level 3 or greater, and 89-98percent for VIP level 2 or greater. PEW varied from 0-2 percent and 0-9 percent for these categories,respectively. As expected, the performance statistics improve with stronger intensity echoes whichoccur less frequently but are more easily distinguishable as true weather or AP. Close examination
7
of data from instances of mis-editing indicated most failures to edit actual AP (i.e., PEAP less than100 percent) involved very small areas of residual breakthrough, while editing of actual weather(i.e., non-zero PEW) involved mainly VIP level 2 returns that were typically on the edge of a precipitation region. Figure 2 is a side-by-side comparison of unedited and AP-edited images that provides an example of the latter occurrence. An overall qualifier of the WSP AP-Editing Algorithm'sperformance observed with this data set is that all significant AP regions were edited sufficientlyto prevent misinterpretation of residual breakthrough as a plausible storm cell, and that no precipitation cells with cores greater than or equal to VIP level 3 were reduced entirely to VIP level 2 or below.
8
,
Figure 2. Unedited (left) and corresponding AP-Edited data (right) from 7110/92. Note removal ofmost AP-induced echoes within AP-truth region denoted bypink polygon. VIP Level 2 weather echoes approximately 25 km northeast ofradar were also edited, resulting in relatively high PEW for this scan.
5. COMPARATIVE RESULTS
As indicated previously, ITWS will censor AP breakthrough on ASR-9 weather channel datadisplayed on dedicated Situation Displays at supervisors' positions. To obtain a sense of comparative perfonnance, an off-line version of the ITWS AP-Editing Algorithm was run on three of thetest cases used to evaluate WSP AP-Editing results. The three cases chosen were among the mostactive AP periods. To simulate the perfonnance of the ITWS AP-editor, TDWR volume scans were"layered" to serve as the Composite Maximum Reflectivity (comprefl) product needed by the algorithm to perfonn editing (Klingle-Wilson, et aI., 1995). In the layering process, the maximum re-
T flectivity value found within the vertical column above each 0.5 square-kilometer resolution binwas used for the composite. Because the testbed TDWR was often in hazardous scanning mode during these time periods, the scoring had to be restricted to the sector scanned by the radar. This usuallycorresponded to a region spanning 100-150 degrees in azimuth. Furthennore, although the heightof the highest TDWR elevation scan (40-60 degrees) was significantly higher than in the currentNEXRAD (NEXt Generation Weather RADar) scanning strategy (approximately 20 degrees), aTDWR cone-of-silence did exist and required that no editing (or scoring) be done within five kilometers of the TDWR. The results of the ITWS AP-Editing simulation appear in Table 3, and a comparison of these with corresponding WSP AP-Editing results are shown in Table 4.
Table 3.Simulated ITWS AP-Editing Results
PIXELS > VIP Level 3 PIXELS > VIP Level 2
DATE PEAP PEW PEAP PEW
8/2/91 712/719 0/2349 4062/4990 40/10,976
8/9/91 8271 / 10,276 0/2009 26,116/40,720 0/7424
7/17/92 619/619 0/289 2841/3304 0/3052
Table 4.Comparison of ITWS and WSP AP-Editing Results
PEAP (> VIP 3) PEW (> VIP 3) PEAP (> VIP 2) PEW (> VIP 2)DATE ITWS WSP ITWS WSP ITWS WSP ITWS WSP
8/2/91 99% 93% 0% 0.7% 81% 94% 0.4% 3.6%
8/9/91 80% 96% 0% 0% 64% 92% 0% 0.2%
7/17/92 100% 98% 0% N/A 86% 98% 0% N/A
11
These limited results indicate the performance of the WSP AP-Editing algorithm is comparableto that of the ITWS AP-Editor. The ITWS version is clearly more conservative in its editing, as evidenced by frequent PEWs near zero percent. This conservatism is also the cause of lower PEAPsat the VIP 2 intensity level relative to the WSP version. A recent performance evaluation of theITWS AP-Editing algorithm (Klingle-Wilson, et aI., 1995) yielded overall PEAPs of 91 percentfor pixels greater than or equal to VIP level 3 and 80 percent for pixels greater than or equal to VIPlevel 2. The corresponding WSP AP-Editing numbers reported here (Table 2) are 96.1 percent and93.5 percent, respectively. In the Klingle-Wilson evaluation, PEWs were 0 and 1 percent for theabove VIP-level categories, while corresponding WSP AP-Editing PEWs reported here are 0.5 and1.4 percent.
12
6. CONCLUSION AND RECOMMENDATION
The results presented in this report indicate the WSP AP-Editing algorithm can provide a reliable capability for editing anomalous propagation contamination from ASR-9 weather-ehannel reflectivity data. The algorithm takes full advantage of the rapid update rate of the ASR radar in determining what data pixels should be edited, and covers all of the terminal airspace. Scoringcomparisons with the ITWS AP-Editing algorithm indicate the WSP AP-Editing algorithm provides comparable performance in editing strong AP (VIP level 3 or greater) and appears superiorin editing lower intensity AP. The WSP AP-edited data will also be directly available to air trafficcontrollers on their DEDS and BRITE displays.
Figure 3, a high-level architecture for the WSP, indicates that the Radar Data Acquisition(RDA) unit-the subsystem responsible for extraction of necessary signals from the ASR-9 andgeneration of base data fields--can be cleanly separated from subsequent wind shear and storm motion product generation and display. We have recommended that the production WSP be designedso that the RDA can be deployed, standalone, as a six-level weather channel replacement at nonWSP sites. This approach would allow the FAA to resolve the AP problem, as well as other identified deficiencies of the weather channel, at larger airports that do not require the full WSP owingto the deployments ofTDWR. As noted previously, although ITWS will display AP-edited weatherchannel data to ATe supervisors at these airports, radar and tower controllers will not have directaccess to this information. A decision to deploy this configuration could be made on an airport-byairport basis, but would probably be very beneficial, in terms of cost, at many large airports.
RDA RPGASR-9 (CLUTTER (WIND SHEAR DFU...... ...... ....... .............. ...... SUPPRESSION ...... AND STORM ...... (DISPLAYS)
INTERFACESAND BASE DATA MOVEMENT
GENERATION) PRODUCTS)
,,SIX-LEVEL
WEATHER TO
DEDS, BRITE
Figure 3. WSP high-level architecture.
13
GLOSSARY
AP Anomalous Propagation
ASR Airport Surveillance Radar
ATC Air Traffic Control
BRITE Bright Radar Indicator Tower Equipment
.- CDM Clear Day Map
DEDS Data Entry and Display System\!
DZ Precipitation Reflectivity
FAA Federal Aviation Administration
FIR Finite Impulse Response
GSD Geographic Situation Display
ITWS Integrated Terminal Weather System
NAS National Airspace System
NEXRAD NEXt Generation Weather RADar
PEAP Probability of Editing AP
PEW Probability of Editing Weather
RDA Radar Data Acquisition
SNR Signal-to-Noise Ratio
SW Spectrum Width
TDWR Terminal Doppler Weather Radar
UCR Unsatisfactory Condition Report
V Radial Velocity
VIP Video Integrator Processor
WSP Weather System Processor
15
..
REFERENCES
Battan, L.J., Radar Observation ofthe Atmosphere, Chicago, IL: University ofChicago Press, 1973.
Evans, J. and D. Turnbull, "Development of an Automated Windshear Detection System UsingDoppler Weather Radar," IEEE, Vol. 77, No. 11, pp. 1661-1673, 1989.
Klingle-Wilson, D.L., "Integrated Terminal Weather System (ITWS) Demonstration and ValidationOperational Test and Evaluation," MIT Lincoln Laboratory, Lexington, MA, Project ReportATC-234, 13 April 1995.
Klingle-Wilson, D., E. Mann and R. Boldi, "An Algorithm to Remove Anomalous PropagationClutter Returns from ASR-9 Weather Channel Data using Pencil Beam Radar Data," Preprints ofThe 6th International Conference on Aviation Weather Systems, American Meteorological Society,Dallas, TX, January 1995.
Puzzo, D.C., S.W. Troxel, M.A. Meister, M.E. Weber and J.V. Pieronek, "ASR-9 Weather ChannelTest Report," MIT Lincoln Laboratory, Lexington, MA, Project Report ATC-165, DOT/FAAIPS-89/3, 3 May 1989.
Taylor, J.W. and G. Bronins, "Design of a New Airport Surveillance Radar (ASR-9)," IEEE, Vol.73,pp.284-289,1985.
Weber, M.E., "Assessment of ASR-9 Weather Channel Performance: Analysis and Simulation,"MIT Lincoln Laboratory, Lexington, MA, Project Report ATC-138, DOT/FAAlPM-86-16, 31 July1986.
Weber, M.E., "Ground Clutter Processing for Wind Measurements With Airport Surveillance Radars," MIT Lincoln Laboratory, Lexington, MA, Project Report ATC-143, DOT/FAAlPM-87/21,4 November 1987.
Weber, M.E., "Dual-Beam Autocorrelation Based Wind Estimates from Airport Surveillance RadarSignals," MIT Lincoln Laboratory, Lexington, MA, Project Report ATC-167, DOT/FAAlPS-89/5,21 June 1989.
Weber, M.E., "Airport Surveillance Radar (ASR-9) Wind Shear Processor: 1991 Test at Orlando,FL," MIT Lincoln Laboratory, Lexington, MA, Project Report ATC-189, DOT/FAAlNR-9217,1 June 1992.
Weber, M.E. and M.L. Stone, "Low Altitude Wind Shear Detection Using Airport Surveillance Radars," IEEE Aerospace and Electronic Systems Magazine, Vol. 10, No.6, pp. 3-9, 1995.
17